[Use of routine data in rehabilitation research - Part 1: An overview of type, access, quality and data protection].

Abstract

Due to the high costs of primary studies, use of existing data, so-called routine data, can be particularly suitable for answering care-related research questions in rehabilitation. Previous reviews on the use of routine data have focused on acute care within the purview of the statutory health insurance (GKV), but have largely overlooked rehabilitation and other rehabilitation-relevant service providers such as the German Pension Insurance (DRV), the German Statutory Accident Insurance (DGUV) and the Federal Employment Agency (BA). The aim is to provide an overview of the type, access, quality and data protection aspects of routine data in the context of rehabilitation research, based on existing recommendations, results of a selective literature search, and the authors' own experience. Routine data is characterized by a large number of cases, large scope of characteristics and longitudinal documentation over long periods of time. Access to routine data from the German Pension Insurance and the Federal Employment Agency is comparatively low threshold for researchers, whereas this is not yet equally the case for data of other social insurance providers and of rehabilitation clinics. Furthermore, under certain conditions, routine data can be linked with other routine data and with primary data, which can considerably expand the spectrum of possible research applications. In addition to the advantages of routine data, their limitations must also be considered. Routine data were collected for other purposes and only contain characteristics that are required for administration. A prospective study approach with routine data is possible in principle due to the continuous data collection and documentation, but randomized allocation to interventions is not feasible. In addition, the availability, generalizability and quality of data sets and individual variables must be verified. The Health Data Lab at the Federal Institute for Drugs and Medical Devices does not yet provide for the integration of GKV rehabilitation data or the linking of GKV data with data from other rehabilitation-relevant service providers. Data protection aspects must also be considered. When using pseudonymized data from social insurance providers, an application must be submitted by the data holder to the relevant supervisory authorities in accordance with § 75 Social Security Code X.